http://avherald.com/h?article=4bff5d1e
Accident: Skylease Cargo B744 at Halifax on Nov 7th 2018, overran runway on landing
QuoteA Skylease Cargo Boeing 747-400, registration N908AR performing flight GG-4854 from Chicago O'Hare,IL (USA) to Halifax,NS (Canada) with 4 crew, landed on Halifax's runway 14 (length 2350 meters/7700 feet) at 05:06L (09:06Z) but was unable to stop before the end of the runway, overran the end of the runway, went down a slope and through the localizer antenna and came to a stop about 200 meters past the runway end. There were no injuries, the aircraft received substantial damage with all gear collapsed, engines #2 and #3 separated, engines #1 and #4 damaged and creases in the fuselage skin.
(http://www.hoppie.nl/tmp/skylease_b744_n908ar_halifax_181107_2.jpg)
That's a pretty short runway -- 6688' on the glideslope. I wonder why they didn't use RW23/05.
Good question....
as per NOTAM, ILS on RW23 was u/s ... RW05 not suitable because of the wind/gust (see metar below)
Minima for LOC 23 is MDA (H) 1000' (442')
METAR:
CYHZ 070900Z 23013KT 7SM -RA BR
BKN005 OVC013 14/14 A2967 RMK SF5SC3 SLP050 DENSITY ALT 800FT=
Ceiling was really "border line" (500') to perform a NP approach on RW23..... but with this wind/gust on RW14 why to continue and not divert?
Quote from: emerydc8 on Thu, 8 Nov 2018 01:53
That's a pretty short runway -- 6688' on the glideslope. I wonder why they didn't use RW23/05.
The landing weight is unknown. With a supposed MLW of 300T and considering a full crosswind, it seems that according to QRH 2347M is doable with manual braking or autobrake MAX and good braking action.
if ILS 23 was U/S, LOC 23 was not an option.
RNAV VNAV (DH 320) or LNAV (MDH 430) 23 was an option. Do they have those approaches available in the FMS navigation database?
NDB 23 (MDH 430) was another option. Even if this approach is not available in the FMS navigation database, VFR option may be used to overlay with LNAV A/P mode as NDB "ZNS" is in the landing axis.
Maybe a wind gust?
To be tried with PSX
RIP. :( Pleased the crew are OK!
One to be tried in PSX!
I tried the ILS 14 with meteorogical conditions described above. I used A/P + Autobrake MAX (9V-SFF / 300T / VREF+10 167Kt). It's no problem.
I think we don't have all the data (variable weather conditions?...)
Quote from: Cbf on Thu, 8 Nov 2018 18:54
I tried the ILS 14 with meteorogical conditions described above. I used A/P + Autobrake MAX (9V-SFF / 300T / VREF+10 167Kt). It's no problem.
I think we don't have all the data (variable weather conditions?...)
1) Is PSX autobrake too good?
2) Accident flight had a fault or failure?
3) Accident flight crew screwed up?
The above is speculation, but do the performance charts say the 744 can get in at that weight and those conditions on that runway? If yes...
PSX manual states MLW is 296.5 tons (652,000 lbs). 300 tons would be
3.7 tons overweight.
PSX data shows a 0.5% runway upslope. Is this accurate to real-world?
Quote from: United744 on Fri, 9 Nov 2018 02:15
1) Is PSX autobrake too good?
PSX aims at the same groundspeed deceleration rates as the real 744:
• 4 ft/sec2 in position 1
• 5 ft/sec2 in position 2
• 6 ft/sec2 in position 3
• 7.5 ft/sec2 in position 4
• 11 ft/sec2 in MAX AUTO
(The higher rates may not be reached on wet runways.)
Quote from: Hardy Heinlin on Fri, 9 Nov 2018 02:24
Quote from: United744 on Fri, 9 Nov 2018 02:15
1) Is PSX autobrake too good?
PSX aims at the same groundspeed deceleration rates as the real 744:
• 4 ft/sec2 in position 1
• 5 ft/sec2 in position 2
• 6 ft/sec2 in position 3
• 7.5 ft/sec2 in position 4
• 11 ft/sec2 in MAX AUTO
(The higher rates may not be reached on wet runways.)
I wasn't doubting it Hardy! :D
I just tried this - LW was 653.2 k lbs (296.9 tons; very slightly overweight).
I did NOT arm the spoilers, and used AUTOBRAKE 3.
I hand flew the last part of the approach, and delibrately took reverse later than I could (I counted to 2 then deployed max reverse). I selected reverse idle at 80 kts and stowed passing 60 kts (IAS). Auto-throttle was used all the way to touchdown ("IDLE also engages on non-autoland" is checked).
I'm not sure I ran off the end, but certainly went past the red lights at the end at around 20-30 kts. I can still see the lead-in lights for the other end in my windshield. The tip of the aircraft symbol on the ND is juuuust past the end of the runway edge markers.
According to PSX data, I landed 1608 ft from the threshold at 2 kts faster than bugged. Flaps 30, 156 kts. I deliberately didn't try to fly it too well, though it was trimmed from AP disengagement and didn't really require much correction. I tried to fly a poor flare to try and get some floating.
If I used Autobrake 4 or MAX, I think I could have made it, but it would be tighter than I would like.
You can check your position on the Instructor map when you zoom in to the max. If a runway end has a different color than the rest of the runway, then that's a stopway or the area beyond a displaced landing threshold.
P.S.: If you enter this METAR in PSX (just paste & copy it into the METAR text edit field in any of the 7 local zones;
it will automatically select CYHZ as the zone center) ...
CYHZ 070900Z 23013KT 7SM -RA BR BKN005 OVC013 14/14 A2967
... you get a 2 kt tailwind component on the approach to runway 14 (and light rain).
Mag var is 17°W. The true heading of Rwy 14 is 125°. Wind is 230° true north. When the aircraft aligns with the runway after touchdown, the tailwind component increases to 3 kt. You can see this in PSX by comparing the TAS and GS indications on the ND.
Of course, the METAR wind is just an approximation. The direction may vary by some degrees. In that real accident in that moment it might have been a better headwind component, or even a worse tailwind component.
Rwy 32 threshold is 41 ft higher than Rwy 14 threshold.
I wonder what autobrake setting these guys used. I would be surprised if they used max, especially if it was just a tech stop and they planned to launch out of there right away. I've never seen max used on the 744. We kind of viewed it as reserved for emergency situations only (arguably it was). I recently saw it used on the 767 while jumpseating on Atlas. I was impressed. I think we got stopped in less than 3000' but we were pretty light and it was a 767-200.
ATC :
https://www.youtube.com/watch?v=sNrMFiqnc0Q
More pics...
https://www.pprune.org/rumours-news/615165-b744f-off-runway-yhz-2.html#post10304980
Quote from: emerydc8 on Fri, 9 Nov 2018 04:53
I've never seen max used on the 744. We kind of viewed it as reserved for emergency situations only (arguably it was).
I was under the impression that maximum manual braking would get you a better (less) braking distance than max auto braking? If so, in case of an emergency would you not go for manual braking?
Jeroen
Yes, you can outbrake max autobrakes, but I think most guys, when faced with the decision of max autobrakes or max-manual, would select max autobrakes and then use manual only if max wasn't doing the job. But then you are just begging to melt fuse plugs.
Here're some data concerning B744 PAX and FREIGHTER weight.
MLW may reach 666.000Lb / 302T for a freighter (GE, RR and PW).
I tried with 300T. Do we know the real weight?
https://www.boeing.com/resources/boeingdotcom/company/about_bca/startup/pdf/historical/747-400-passenger.pdf
http://www.boeing.com/resources/boeingdotcom/company/about_bca/startup/pdf/freighters/747-400f.pdf
Quote from: emerydc8 on Fri, 9 Nov 2018 17:58
Yes, you can outbrake max autobrakes, but I think most guys, when faced with the decision of max autobrakes or max-manual, would select max autobrakes and then use manual only if max wasn't doing the job. But then you are just begging to melt fuse plugs.
I also think that all else being equal, MAX AUTO will do a better job getting the airplane stopped in a STABLE way, balancing left and right better than a human can who is stamping on the pedals as hard as (s)he can and in the mean time tries to accomplish everything else to stop the beast.
Hoppie
Actually, the distance is significant. On the 767, AB max at 326,000 lbs at sea level on a dry runway is 5184' versus 3132' for max manual braking.
I agree with you that autobrakes are better at applying equal pressure on each brake, but also the tendency with manual braking is that the brakes will likely be over-applied and then backed off, then over-applied, etc. The AB system does a much better job of continuous application and actually backs off the brake pressure as the speed decreases.
Is the antiskid system of no help when braking manually?
(Antiskid failures in PSX has some consequences ...)
QuoteIs the antiskid system of no help when braking manually?
Absolutely, but you're probably going to pay for it with hot brakes if max manual is applied. When I say "over-applied," I mean within the limits of the anti-skid. So, all this braking is going on without the anti-skid getting involved. We just can't do as good a job here. I have to admit.
You'd be surprised what an extra second or two of slightly favoring one side will do to the temps. It could mean an extra few hundred degrees just for doing it while exiting a high-speed, at least on our steel-brake aircraft. When that happens, I'll try not to use that brake during taxi-in until both sides are about the same temp. I'll just counter the pull with tiller.
(EDIT: Sorry.. cross-posted)
QuoteIs the antiskid system of no help when braking manually?
I think Jon's comments are simply highlighting the limitations of human braking. Max manual brakes feels severe, so the pilots back off... then realise they have backed off too much.. and reapply max braking, repeating the cycle.. or may be hestitant to reapply max braking. Or, the pilots apply unequal pressure on the left and right pedals due to leg strength/length, so have to figure out the correct balance to stop them running off the side of the runway. The antiskid and torque-limiting systems are always in operation.
The autobrakes provide relatively smooth application and
balanced left/right application pressure automatically. The braking should be straight (unless the wheels on one side of the aircraft cannot get a grip at all due to long sections of ice/deep water only on that side). By balanced, I mean there is only one outlet pipe on the autobrake pressure control module. This hydraulic pressure goes through "T" pieces (pipes) to the normal antiskid valves for all bogeys. The pilot can apply different pressures to each side.
"Balanced left/right application pressure" -- does this aim at equal pressure or at avoiding any yaw? And if it's the latter, what are the control criteria? Sensed IRS yaw or sensed wheel speeds?
QuoteMax manual brakes feels severe, so the pilots back off... then realise they have backed off too much.. and reapply max braking, repeating the cycle.. or may be hestitant to reapply max braking.
+1
Quote
"Balanced left/right application pressure" -- does this aim at equal pressure or at avoiding any yaw? And if it's the latter, what are the control criteria? Sensed IRS yaw or sensed wheel speeds?
I've found the AB to do a great job keeping it on centerline, even if there's a momentary release on one side going over a patch of ice.
I tend to favor the right brake because I can rest my right knee up against the center console and I have a reference point for braking. If you play the guitar, it's like resting the edge of your your picking hand on the bridge to have more control. So for me, if one side is hot, it's usually the right side.
I'm left-foot dominant. I'll lead with the left foot and balance with the right. If any brake needs to be modulated to straighten things out, I'll modulate the right foot to match the left once I've got the braking effort I desire.
Never thought about it from that point of view. I think I'm left-foot dominant too, but I have a heavy right foot for reasons explained above.
Quote"Balanced left/right application pressure" -- does this aim at equal pressure or at avoiding any yaw? And if it's the latter, what are the control criteria? Sensed IRS yaw or sensed wheel speeds?
No, by application pressure, I mean input pressure
into the antiskid valves. What comes out of the antiskid valves will be
unbalanced, depending on which wheels are skidding.
QuoteI've found the AB to do a great job keeping it on centerline, even if there's a momentary release on one side going over a patch of ice.
Perhaps (as I think was previously discussed in other threads), inertia is playing a key part here. At higher speeds, the aircraft has a tendency to travel in a straighter line.... Probably why even the unbraked nose gear tends to skid at higher speeds when it's used prematurely.
Quote from: John H Watson on Sat, 10 Nov 2018 04:19
No, by application pressure, I mean input pressure into the antiskid valves.
I thought you meant the
autobrakes, not the antiskid:
QuoteThe autobrakes provide relatively smooth application and balanced left/right application pressure ...
QuoteI thought you meant the autobrakes, not the antiskid:
I did mean the autobrakes. The autobrakes provide a controlled/metered hydraulic pressure (instead of the pilot) and it goes to all 4 normal antiskid modules and then to the brakes. The antiskid modules interfere with this pressure flow if a skid is detected.
hydraulic pumps > autobrake module > all antiskid modules > all brakes.
With manual braking, the pilots modulate the brake pedals which controls the flow of hydraulic fluid to the left and right pairs of antiskid modules.
hydraulic pumps > left/right brake pedal-operated metering valves > L/R antiskid module pairs > L/R brakes
I see. But it's the antiskid which maintains the aircraft track when an asymmetric skid is detected? (The antiskid doesn't "steer", but thanks to its asymmetric skid detection the track is pretty stable.) The metered pressure of the autobrakes is controlled by the IRS groundspeed (or wheelspeed if IRS invalid), and does not try to maintain the track by asymmetric braking, correct?
QuoteBut it's the antiskid which maintains the aircraft track when an asymmetric skid is detected?
No, the antiskid or the autobrake system doesn't have asymmetry protection. The only crosstalk between the brakes is between the wheels of the same bogey under certain conditions. This cannot help with balancing the left side with the right side. I think only the inertia of the aircraft keeps the aircraft going in a straight line (up to a point).
Perhaps if the plane (slowly) veered to the side of the runway and two wing gear wheels were on the grass, the aircraft would brake more on the opposite side and pull you back onto the runway... but if a complete bogey started to sink into the grass, the drag alone might keep you in the grass.
Wheelspeed cannot replace the IRU input. IRU failure will trip the A/B switch. Wheelspeed is only used for application logic.
QuoteFlat main gear tire(s) cause a general loss of braking effectiveness and a yawing moment toward the flat tire with light or no braking and a yawing moment away from the flat tire if the brakes are applied harder. Maximum use of reverse thrust is recommended. Do not use autobrakes.
B747 FCTM 6.24
The aircraft could stop in that distance. No question about that. However given the nature of the gusty surface wind it would be difficult to get the aircraft down in the correct place, and apply adequate braking.
Manual braking should exceed the autobrake capabilities but it depends on how it is used. Dealing with a crosswind yaw AND applying maximum manual braking is extremely difficult. If they had set Max Auto or even level 4 the braking will not go to the stated rate until the pitch angle is below a certain amount anyway (to prevent the brakes crunching the nose in).
It is very tempting to try and brake differentially to help with directional control. However there is a real downside to this. As soon as you release brake pressure on a pedal that brake system will go through a slight delay before the brakes take effect again because they go through their anti skid sensing cycle again - the only way to beat this is to apply more pressure on the other brake - but if you already have full pressure applied - that won't work..
Also it's actually quite difficult to apply rudder for directional control without disengaging the autobrakes accidentally. This is not uncommon on significant crossword landings. If you don't spot this straight away you will have lost valuable distance..
Just a few possibilities.
Lots going on here. I can't help but feel the CVR will be very interesting with regard to CRM..
We shouldn't be pointing fingers really, but - hey - as no-one was hurt..